Skip to main content

Table 3 Deep neural network

From: Recognition of cooking activities through air quality sensor data for supporting food journaling

 

Original predictions

After post-processing

TN

296,219

301,258

FP

37,978

32,939

TP

12,055

11,985

FN

4269

4339

Accuracy (%)

87.95

89.36

Specificity (%)

88.64

90.14

Recall (%)

73.85

73.42

Precision (%)

24.09

26.68

\(F_1\) score (%)

36.33

39.14

  1. Results with minute by minute modality, online recognition
  2. Numbers in italic indicate the largest value obtained in the experiment for each considered metric
Back to article page